A spectrum analysis method, system, storage medium and computer device for forging quality detection

Abstract

The present application relates to the technical field of forging spectral data analysis, and particularly relates to a spectral analysis method and system for forging quality detection, a storage medium and computer equipment. After obtaining forging spectral data, the present application selects reference forgings, divides wave bands based on spectral curve adjacent wavelength data difference and wave peak position, and extracts atomic oxygen and auxiliary peak wave bands. After selecting target wave bands, the present application obtains oxidation degree and weight by comparing normal forging calculation wave peak difference and data difference, and obtains final oxidation degree by combining weight correction. The present application analyzes component wave band change state, corrects spectral data by combining oxidation degree, and completes quality detection. The present application can eliminate the influence of oxidation film on forging spectral detection, so as to collect accurate forging spectral data, and further help relevant personnel to obtain more accurate forging quality detection results.

Description

Technical Field

[0001] This invention relates to the technical field of spectral data analysis of forgings, and specifically to a spectral analysis method, system, storage medium, and computer equipment for forging quality inspection. Background Technology

[0002] With the rapid development of modern industry towards intelligence and high precision, forgings, as core components in key fields such as aerospace, energy equipment, and rail transportation, face increasingly stringent quality requirements. Traditional testing methods (such as destructive mechanical testing and X-ray flaw detection) suffer from low efficiency, high cost, or inability to provide real-time feedback, making it difficult to meet the demands of intelligent manufacturing for online and non-destructive testing. Spectroscopic analysis technologies (such as LIBS and OES), with their advantages of non-contact and simultaneous multi-element analysis, have become an important direction for breaking through traditional bottlenecks. This technology, by integrating artificial intelligence algorithms with portable equipment design, can not only achieve rapid and accurate diagnosis of forging composition and defects but also link with the production line in real time to optimize process parameters and reduce scrap rates and energy consumption. Its development is of significant importance for improving the reliability of high-end equipment and promoting the green upgrading of manufacturing, and is a key link in the transition of quality inspection from "post-event sampling" to "full-process intelligent monitoring."

[0003] When using spectral technology to inspect the quality of forgings, oxidation may occur during the forging process, resulting in an oxide film covering the surface of the forging. After pretreatment, a thin oxide film may remain on the surface of the forging, which will affect the data collected by spectral detection of the forging, thus causing deviations in the collected spectral data and abnormal forging quality inspection results. Summary of the Invention

[0004] To address the issue that oxidation may occur during the forging stage of forging production, resulting in an oxide film covering the forging surface, and even after pretreatment, a thin oxide film may remain on the forging surface, this present invention aims to provide a spectral analysis method, system, storage medium, and computer equipment for forging quality inspection. The specific technical solution adopted is as follows:

[0005] A spectral analysis method for forging quality inspection, the method comprising:

[0006] Obtain the spectral data of all forgings for each forging;

[0007] Select any forging as a reference forging; obtain the spectral data curves corresponding to the spectral data of all forgings on the reference forging; divide the spectral data curves into bands based on the data differences between the spectral data of forgings at adjacent wavelengths and the peak positions on the spectral data curves to obtain all bands on the spectral data curves; obtain the atomic oxygen band and auxiliary peak band from all bands; select either the atomic oxygen band or the auxiliary peak band as the target band; obtain the oxidation degree of the target band of the reference forging based on the peak differences between the reference forging and normal forgings on the target band, and the data differences of the reference forging on the target band; obtain the oxidation weight of the target band of the reference forging based on the curve differences between the reference forging and normal forgings on the target band; correct the oxidation degree of the atomic oxygen band and the auxiliary peak band based on the oxidation weights in the atomic oxygen band and the auxiliary peak band to obtain the final oxidation degree of the reference forging;

[0008] Obtain the wavebands corresponding to all components in the reference forging; obtain the change state of the forging spectral data corresponding to each component; correct the forging spectral data of each component according to the change state and the final oxidation degree to obtain the final spectral data; and perform quality inspection on the reference forging according to the final spectral data.

[0009] Furthermore, the methods for obtaining all bands on the spectral data curve include:

[0010] Select any one of the reference forgings as the reference data;

[0011] Calculate the data difference between the reference data and the forging spectral data corresponding to the previous wavelength;

[0012] Obtain all extreme points of the spectral data curve corresponding to the reference forging;

[0013] The curve segment between each two adjacent extreme points is used as the initial waveband;

[0014] The peak value of the forging spectral data corresponding to the previous wavelength of the reference data in the initial band is taken as the temporary peak.

[0015] Based on the positional distribution of the reference data and the temporary peaks, and the data difference between the reference data and the forging spectral data corresponding to the previous wavelength, the probability that the reference data and the previous forging spectral data are classified into the same initial wavelength band is obtained, as shown in the following formula:

[0016] ;

[0017] In the formula, Indicates the first The spectral data of the first forging and the first The likelihood that the spectral data of each forging is divided into the same initial band; This represents the distance from the first forging spectral data in the initial band to the temporary peak; Indicates the time from the temporary peak to the first peak. The distance between the spectral data of each forging; Indicates the first The spectral data of the first forging and the first Data differences between the spectral data of individual forgings; Represents an exponential function with the natural constant as its base; Represents the normalization function;

[0018] when When it exceeds the preset first threshold, the first... The spectral data of the forging was divided into the first... The initial band where the spectral data of each forging is located, when When it is less than the preset first threshold, the first The spectral data of one forging is used as the starting spectral data of the next band; all spectral data of forging are traversed, and the spectral data of all forging are divided into bands to obtain all bands on the spectral data curve.

[0019] Furthermore, the method for obtaining the degree of oxidation includes:

[0020] The degree of oxidation is obtained according to the formula for calculating the degree of oxidation, which is shown below:

[0021] ;

[0022] In the formula, Indicates the target band number; Indicates the degree of oxidation in the target wavelength band; Indicates the peak value of the target waveband of the reference forging; Indicates the peak value of the target waveband of the forging under normal conditions; This indicates the difference between the peak wavelengths of a forging under normal conditions and a reference forging in the target band. This represents the normalization function.

[0023] Furthermore, the method for obtaining the oxidation weight includes:

[0024] The oxidation weight is obtained according to the oxidation weight calculation formula, which is shown below:

[0025] ;

[0026] In the formula, Indicates the target band number; This indicates the number of wavelengths that overlap between the forging and the reference forging under normal conditions in the target band. This indicates the first superposition of normal forgings in the target band. Spectral data of forgings corresponding to each wavelength; This indicates the first superposition of the reference forging in the target band. Spectral data of forgings corresponding to each wavelength; This represents the normalization function.

[0027] Furthermore, the method for obtaining the final degree of oxidation includes:

[0028] The final oxidation degree is obtained according to the formula for calculating the final oxidation degree, which is as follows:

[0029] ;

[0030] ;

[0031] In the formula, Indicates the target band number; This indicates the overall weight of the target band; Indicates the oxidation weight of the target band; This indicates the oxidation weight of the other band besides the target band in the atomic oxygen band and the auxiliary peak band; Indicates the final oxidation degree of the reference forging; Indicates the degree of oxidation in the target wavelength band; This indicates the degree of oxidation in the atomic oxygen band and the auxiliary peak band, in addition to the target band.

[0032] Furthermore, the method for obtaining the final spectral data includes:

[0033] The final spectral data is obtained according to the final spectral data calculation formula, which is as follows:

[0034] ;

[0035] In the formula, Indicates the first in the reference forging Of the components, the first The corresponding final spectral data for each forging spectral data; Indicates the first in the reference forging The first of the components Spectral data of individual forgings; Indicates the first The variation state of the spectral data of the forgings corresponding to each component; This indicates the final oxidation degree of the reference forging.

[0036] A spectral analysis system for forging quality inspection includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the spectral analysis method for forging quality inspection as described above.

[0037] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the spectral analysis method for forging quality inspection as described above.

[0038] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of the spectral analysis method for forging quality inspection as described above.

[0039] The present invention has the following beneficial effects:

[0040] This invention acquires all spectral data of each forging. Due to the influence of the oxide film on the forging, the collected spectral data of the forging at each wavelength will vary to a certain extent, thus affecting the band division on the spectral data curve. Therefore, the spectral data curve is divided into bands by the data differences between adjacent wavelengths of the forging spectral data and the peak positions on the spectral data curve, thus obtaining all bands on the spectral data curve. Since the spectral data of the bands corresponding to the atomic oxygen characteristic peak and auxiliary peak can reflect the coverage of the oxide film on the surface of the forging to a certain extent, the atomic oxygen band and auxiliary peak band are acquired from all bands. After the forging is processed by the rotating wire brush, the oxide film on the surface is in a state of low thickness. At this time, the degree of oxidation on the surface of the forging will directly affect the spectral data of the atomic oxygen band and auxiliary peak band. Therefore, one of the atomic oxygen band and auxiliary peak band is selected as the target band. Based on the target band of the reference forging and the normal forging, the spectral data of the reference forging and the normal forging are divided into bands. The oxidation degree of the reference forging in the target band is obtained by analyzing the peak differences in the band and the data differences of the reference forging in the target band. Since the atomic oxygen band and auxiliary peak band induce oxidation to different degrees on different forgings, the oxidation weight of the reference forging in the target band is obtained based on the curve differences between the reference forging and normal forgings. This oxidation weight is then used to correct the oxidation degree to obtain the final oxidation degree of the forging. Because there are many components in the forging, the spectral data corresponding to different components are affected differently in the presence of the oxide film. Therefore, the bands corresponding to all components in the reference forging are obtained. Based on the characteristic relationship between each component and the oxide film, the change state of the forging spectral data corresponding to each component in the forging is obtained when the oxide film on the forging surface thickens. Based on the change state and the final oxidation degree, the forging spectral data of each component is corrected to obtain the final spectral data. The reference forging is then subjected to quality inspection based on the final spectral data. This invention can eliminate the influence of the oxide film on the spectral detection of forgings, thereby collecting accurate forging spectral data and helping relevant personnel obtain more accurate forging quality inspection results. Attached Figure Description

[0041] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0042] Figure 1 A flowchart of a spectral analysis method for forging quality inspection provided in one embodiment of the present invention;

[0043] Figure 2This is a flowchart of a spectral analysis system for forging quality inspection provided in one embodiment of the present invention. Detailed Implementation

[0044] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a spectral analysis method, system, storage medium, and computer device for forging quality inspection proposed according to the present invention. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0045] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0046] The following detailed description, in conjunction with the accompanying drawings, illustrates a specific scheme for a spectral analysis method, system, storage medium, and computer equipment for forging quality inspection provided by the present invention.

[0047] Please see Figure 1 This illustrates a spectral analysis method for forging quality inspection provided by an embodiment of the present invention, the method comprising:

[0048] Step S1: Obtain all forging spectral data for each forging.

[0049] This invention is primarily applied to the detection of oxide films on the surface of forgings. Because this invention utilizes spectroscopy for forging analysis, it acquires spectral data for all forgings for each individual forging.

[0050] In this embodiment of the invention, a method for acquiring spectral data of forgings is provided, specifically including:

[0051] First, the forging inspection and fixing module uses a robotic arm to grip the forgings present in the production line with constant pressure and position them at the inspection station. Simultaneously, compressed air is used to blow away surface oil and dust, and a rotating wire brush polishes the inspection points, roughly removing the oxide layer on the forging surface. In one embodiment of the invention, the compressed air blowing pressure is set to 0.6-0.8 MPa, the nozzle is 50 mm from the forging surface, and the rotating wire brush has a diameter of 30 mm and a rotation speed of 3000 rpm.

[0052] Then, a laser emitter is used to emit pulses for laser excitation (power: 50-100mJ, focused spot diameter: 100um). The emission spectrum is simultaneously acquired by a spectrometer (acquisition delayed by 1us after laser triggering, integration time: 1ms), with a wavelength range of 200-900nm.

[0053] Perform full-coverage spectral detection of the forging surface according to a preset grid (such as the preset grid). (Dot matrix) The collected spectral data is preprocessed to obtain the forging spectral data required for subsequent analysis steps. The forging spectral data is in the form of (wavelength, emitted light intensity).

[0054] It should be noted that in other embodiments of the present invention, the method for obtaining the spectral data of the forging can be set by the user and is not limited here.

[0055] Step S2: Select any forging as a reference forging; obtain the spectral data curves corresponding to the spectral data of all forgings on the reference forging; divide the spectral data curves into bands based on the data differences between the spectral data of forgings at adjacent wavelengths and the peak positions on the spectral data curves, obtaining all bands on the spectral data curves; obtain the atomic oxygen band and auxiliary peak band from all bands; select either the atomic oxygen band or the auxiliary peak band as the target band; obtain the oxidation degree of the target band of the reference forging based on the peak differences between the reference forging and normal forgings on the target band, and the data differences of the reference forging on the target band; obtain the oxidation weight of the target band of the reference forging based on the curve differences between the reference forging and normal forgings on the target band; correct the oxidation degree of the atomic oxygen band and auxiliary peak band based on the oxidation weights in the atomic oxygen band and auxiliary peak band, obtaining the final oxidation degree of the reference forging.

[0056] In reality, the oxide film on the surface of forgings can affect the accuracy of the collected spectral data, impacting subsequent analysis and the accuracy of forging quality inspection results. Firstly, due to the inherent properties of forgings, they contain numerous components, resulting in a wide range of wavelengths in their spectral data curves. The oxide film causes variations in the collected spectral data at different wavelengths, affecting the band division on the spectral data curve. Therefore, a Cartesian coordinate system is established with wavelength as the horizontal axis and emitted light intensity as the vertical axis. The acquired forging spectral data is placed within this system, and all spectral data from a reference forging are fitted to a single spectral data curve. By analyzing the differences between spectral data at adjacent wavelengths and the peak positions on the spectral data curve, band division is performed to obtain all bands on the spectral data curve.

[0057] Preferably, in one embodiment of the present invention, the method for obtaining all bands on the spectral data curve includes:

[0058] Select any forging spectral data from the reference forgings as the reference data; calculate the data difference between the reference data and the forging spectral data corresponding to the previous wavelength; obtain all extreme points of the spectral data curve corresponding to the reference forging; take the curve segment between each two adjacent extreme points as the initial band. Obtain the peak value in the initial band of the forging spectral data corresponding to the previous wavelength of the reference data as the temporary peak.

[0059] Based on the positional distribution of the reference data and the temporary peaks, and the data difference between the reference data and the forging spectral data corresponding to the previous wavelength, the probability that the reference data and the previous forging spectral data are classified into the same initial wavelength band is obtained, as shown in the following formula:

[0060] ;

[0061] In the formula, Indicates the first The spectral data of the first forging and the first The likelihood that the spectral data of each forging is divided into the same initial band; This represents the distance from the first forging spectral data in the initial band to the temporary peak; Indicates the time from the temporary peak to the first peak. The distance between the spectral data of each forging; Indicates the first The spectral data of the first forging and the first Data differences between the spectral data of individual forgings; Represents an exponential function with the natural constant as its base; This represents the normalization function.

[0062] In the probability calculation formula, the smaller the difference between the spectral data of forgings corresponding to adjacent wavelengths, the better. The spectral data of the first forging and the first The smaller the difference in reflected light intensity of the spectral data of each forging, the better if the first... The greater the difference in distance between the spectral data of each forging and the spectral data of the first forging in the initial band, and the temporary peak, the better. The larger it is, the more likely it is to be the first The more likely the spectral data of a forging is to also fall within that initial band, that is, at this time... The smaller the hour, the more... The more likely the spectral data of a forging is to also fall within that initial band, the more likely it is to indicate that the first... The spectral data of the first forging and the first The greater the likelihood that the spectral data of each forging will be divided into the same initial band.

[0063] when When it exceeds the preset first threshold, the first... The spectral data of the forging was divided into the first... The initial band where the spectral data of each forging is located, when When it is less than the preset first threshold, the first The spectral data of each forging is used as the starting spectral data for the next band; all spectral data of forgings are traversed, and bands are divided for all spectral data to obtain all bands on the spectral data curve. In one embodiment of the present invention, the preset first threshold is set to 0.7, and can be set by the user, which is not limited here.

[0064] Since the spectral data of the characteristic peaks of atomic oxygen and the bands corresponding to the auxiliary peaks can reflect the degree of oxide film coverage on the surface of the forging to a certain extent, in this embodiment of the invention, the atomic oxygen band and the auxiliary peak band are obtained from all bands.

[0065] In one embodiment of the present invention, since the characteristic peak of atomic oxygen in the forging is generally located at a wavelength of 777.2 nm, and the auxiliary peak is generally located at 844.6 nm, the band with the wavelength closest to 777.2 nm corresponding to the peak is selected as the atomic oxygen band from the bands divided from the spectral data curve of the forging; similarly, the band with the wavelength closest to 844.6 nm corresponding to the peak is selected as the auxiliary peak band from the divided bands.

[0066] After the forgings are treated with a rotating wire brush, the oxide film on their surface is relatively thin. At this point, the degree of oxidation on the forging surface directly affects the spectral data in the atomic oxygen band and the auxiliary peak band. Therefore, we first analyze forgings in their normal state, where no oxidation has occurred. We then select either the atomic oxygen band or the auxiliary peak band as the target band. Based on the peak differences between the reference forging and the normal forging in the target band, as well as the data differences of the reference forging in the target band, we obtain the degree of oxidation of the reference forging in the target band.

[0067] It should be noted that the calculation method for the degree of oxidation in the atomic oxygen band and the auxiliary peak band is the same, therefore, this embodiment of the invention only describes the calculation method.

[0068] Preferably, in one embodiment of the present invention, the method for obtaining the degree of oxidation includes:

[0069] The degree of oxidation is obtained according to the formula for calculating the degree of oxidation, which is shown below:

[0070] ;

[0071] In the formula, Indicates the target band number; Indicates the degree of oxidation in the target wavelength band; Indicates the peak value of the target waveband of the reference forging; Indicates the peak value of the target waveband of the forging under normal conditions; This indicates the difference between the peak wavelengths of a forging under normal conditions and a reference forging in the target band. This represents the normalization function.

[0072] In the formula for calculating the degree of oxidation, the greater the difference in peak value between the reference forging and the normal forging, and the greater the change in the wavelength corresponding to the peak value, the greater the influence of the oxide film on the forging spectral data, indicating that the degree of oxidation in the target band is greater.

[0073] Thus, the degree of oxidation in the atomic oxygen band and the auxiliary peak band has been calculated.

[0074] Since the degree of oxidation caused by the atomic oxygen band and the auxiliary peak band varies on different forgings, the oxidation weight of the target band of the reference forging is obtained based on the difference in the curves of the reference forging and the normal forging on the target band.

[0075] Preferably, in one embodiment of the present invention, the method for obtaining the oxidation weight includes:

[0076] The oxidation weight is obtained according to the oxidation weight calculation formula, which is shown below:

[0077] ;

[0078] In the formula, Indicates the target band number; This indicates the number of wavelengths that overlap between the forging and the reference forging under normal conditions in the target band. This indicates the first superposition of normal forgings in the target band. Spectral data of forgings corresponding to each wavelength; This indicates the first superposition of the reference forging in the target band. Spectral data of forgings corresponding to each wavelength; This represents the normalization function.

[0079] In the oxidation weighting calculation formula, the differences in spectral data between the atomic oxygen band and the auxiliary peak band reflect the thickness of the oxide film on the forging surface. When the overall difference in a certain band is large, it reflects a greater degree of oxidation on the forging surface, thus giving it a larger weight value when calculating the degree of oxidation on the forging surface. The overlap between the normal forging and the reference forging in the target band... The greater the difference between the spectral data of the forging corresponding to each wavelength, and the fewer the number of spectral data that can overlap, the greater the difference in the spectral data of the forging in the target band. In this case, the target band should be given a greater weight.

[0080] Preferably, in one embodiment of the present invention, the method for obtaining the final degree of oxidation includes:

[0081] The final oxidation degree is obtained according to the formula for calculating the final oxidation degree, which is shown below:

[0082] ;

[0083] ;

[0084] In the formula, Indicates the target band number; This indicates the overall weight of the target band; Indicates the oxidation weight of the target band; This indicates the oxidation weight of the other band besides the target band in the atomic oxygen band and the auxiliary peak band; Indicates the final oxidation degree of the reference forging; Indicates the degree of oxidation in the target wavelength band; This indicates the degree of oxidation in the atomic oxygen band and the auxiliary peak band, in addition to the target band.

[0085] In the formula for calculating the final degree of oxidation, firstly using The oxidation weights are normalized; if the target band is the atomic oxygen band, then... Obtain the comprehensive weight of the auxiliary peak band; multiply the comprehensive weight of each of the two bands by the corresponding oxidation degree and add them together to obtain the final weight. The final oxidation degree of the forging is used as a reference.

[0086] Step S3: Obtain the wavebands corresponding to all components in the reference forging; obtain the change state of the forging spectral data corresponding to each component; correct the forging spectral data of each component according to the change state and the final oxidation degree to obtain the final spectral data; perform quality inspection on the reference forging according to the final spectral data.

[0087] Based on the oxidation degree of the forging surface analyzed above, and combined with the spectral data of each band of the forging itself, the collected spectral data of the forging are corrected to reduce the influence of the oxide film on the spectral detection data, making the subsequent results of forging quality inspection more accurate and reliable.

[0088] Because forgings contain a variety of components, the spectral data corresponding to different components are affected to varying degrees in the presence of an oxide film. For example, the characteristic peaks of light elements such as C, S, and P decrease with increasing oxide film thickness. Therefore, in this embodiment of the invention, the wavebands corresponding to all components in the reference forging are obtained.

[0089] In one embodiment of the present invention, the peaks of each band divided in the forging are compared with the theoretically corresponding peaks of each component, and the band corresponding to the nearest peak is selected as the band corresponding to each component. It should be noted that the method for obtaining the band corresponding to each component is a technique well-known to those skilled in the art and will not be elaborated upon here.

[0090] Based on the characteristic relationship between each component and the oxide film, the change state of the forging spectral data corresponding to each component in the forging is obtained as the oxide film on the surface of the forging thickens. It should be noted that the change state is a technique well known to those skilled in the art, and will not be elaborated here.

[0091] It should be noted that when the change state is increasing, its value is 1, and when the change state is decreasing, its value is -1. That is, when the change state is 1, it means that when the surface of the forging is thickened, the spectral data of the forging corresponding to this component will increase, and when the change state is -1, it means that when the surface of the forging is thickened, the spectral data of the forging corresponding to this component will decrease.

[0092] Based on the change state and the final oxidation degree, the spectral data of the forgings for each component are corrected to obtain the final spectral data.

[0093] Preferably, in this embodiment of the invention, the method for obtaining the final spectral data includes:

[0094] The final spectral data is obtained according to the calculation formula, which is shown below:

[0095] ;

[0096] In the formula, Indicates the first in the reference forging Of the components, the first The corresponding final spectral data for each forging spectral data; Indicates the first in the reference forging The first of the components Spectral data of individual forgings; Indicates the first The variation state of the spectral data of the forgings corresponding to each component; This indicates the final oxidation degree of the reference forging.

[0097] In the final spectral data calculation formula, the greater the final oxidation degree of the forging surface, the greater the influence on the spectral data of each component of the forging at this time, and the greater the degree of correction. At the same time, the correction process must conform to the change state characteristics of each component of the forging.

[0098] Based on the corrected final spectral data, the content of each component in the forging can be analyzed, thereby aiding in the quality inspection of the forging. Specifically:

[0099] First, it is necessary to build a model between spectral data and component content. The training set is composed of spectral data collected from multiple precisely processed post-forging parts and the content of each component detected manually, which helps to build the model between spectral data and component content.

[0100] The final spectral data is input into the trained model to obtain the content of each component in the current forging. This content is then compared with the standard content of each component in the forging to identify any abnormalities, thus completing the quality inspection of the forging.

[0101] In summary, the following steps are taken: First, obtain the spectral data of all forgings for each forging. Then, select one forging as a reference forging. Next, obtain the spectral data curves corresponding to the spectral data of all forgings on the reference forging. Based on the data differences between the spectral data of forgings at adjacent wavelengths and the peak positions on the spectral data curves, divide the spectral data curves into bands to obtain all bands on the spectral data curves. Then, obtain the atomic oxygen band and the auxiliary peak band from all bands. Finally, select either the atomic oxygen band or the auxiliary peak band as the target band. Based on the peak differences between the reference forging and normal forgings in the target band, and the data differences of the reference forging in the target band, obtain... The oxidation degree of the reference forging in the target band is obtained; based on the curve difference between the reference forging and the normal forging in the target band, the oxidation weight of the reference forging in the target band is obtained; based on the oxidation weight in the atomic oxygen band and the auxiliary peak band, the oxidation degree of the atomic oxygen band and the auxiliary peak band is corrected to obtain the final oxidation degree of the reference forging; the bands corresponding to all components in the reference forging are obtained; the change state of the forging spectral data corresponding to each component is obtained; based on the change state and the final oxidation degree, the forging spectral data of each component is corrected to obtain the final spectral data; the reference forging is quality inspected based on the final spectral data.

[0102] A second objective of one embodiment of the present invention is to provide a spectral analysis system for forging quality inspection. This system includes a memory, a processor, and a computer program. The memory stores the corresponding computer program, and the processor runs the corresponding computer program. When the computer program runs in the processor, it can implement the methods described in steps S1-S3, specifically including:

[0103] Data acquisition module 101 is used to acquire the spectral data of all forgings for each forging;

[0104] The data analysis module 102 is used to select any forging as a reference forging; obtain the spectral data curves corresponding to the spectral data of all forgings on the reference forging; divide the spectral data curves into bands according to the data differences between the spectral data of forgings at adjacent wavelengths and the peak positions on the spectral data curves, and obtain all bands on the spectral data curves; obtain the atomic oxygen band and auxiliary peak bands from all bands; select one of the atomic oxygen band and auxiliary peak bands as the target band; obtain the oxidation degree of the target band of the reference forging according to the peak differences between the reference forging and normal forgings on the target band, and the data differences of the reference forging on the target band; obtain the oxidation weight of the target band of the reference forging according to the curve differences between the reference forging and normal forgings on the target band; and correct the oxidation degree of the atomic oxygen band and auxiliary peak bands according to the oxidation weights in the atomic oxygen band and auxiliary peak bands to obtain the final oxidation degree of the reference forging.

[0105] The quality inspection module 103 is used to acquire the wavebands corresponding to all components in the reference forging; obtain the change state of the forging spectral data corresponding to each component; correct the forging spectral data of each component according to the change state and the final oxidation degree to obtain the final spectral data; and perform quality inspection on the reference forging according to the final spectral data.

[0106] A third objective of this invention is to provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the methods described in steps S1-S3.

[0107] The fourth objective of this invention is to provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the method described in steps S1-S3.

[0108] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0109] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A spectral analysis method for forging quality detection, characterized in that, The method comprises: Obtaining all forging spectral data of each forging; Optionally, one forging is taken as a reference forging; obtaining the spectral data curve corresponding to all forging spectral data on the reference forging; dividing the spectral data curve into all wave bands according to the data difference between the forging spectral data of adjacent wavelengths and the position of the wave peak on the spectral data curve; obtaining all wave bands on the spectral data curve; obtaining the atomic oxygen wave band and the auxiliary peak wave band in all wave bands; optionally taking one of the atomic oxygen wave band and the auxiliary peak wave band as a target wave band; obtaining the oxidation degree of the target wave band of the reference forging according to the wave peak difference of the target wave band between the reference forging and the normal forging and the data difference of the target wave band of the reference forging; obtaining the oxidation weight of the target wave band of the reference forging according to the curve difference of the target wave band between the reference forging and the normal forging; correcting the oxidation degree of the atomic oxygen wave band and the auxiliary peak wave band according to the oxidation weight in the atomic oxygen wave band and the auxiliary peak wave band to obtain the final oxidation degree of the reference forging; Obtaining the wave band corresponding to all components in the reference forging; obtaining the change state of the forging spectral data corresponding to each component; correcting the forging spectral data of each component according to the change state and the final oxidation degree to obtain the final spectral data; and performing quality detection on the reference forging according to the final spectral data; The method for obtaining the final oxidation degree comprises: The final oxidation degree is obtained according to the final oxidation degree calculation formula, and the final oxidation degree calculation formula is as follows: In the formula, denotes the serial number of the target band; denotes the comprehensive weight of the target band; denotes the oxidation weight of the target band; denotes the oxidation weight of another band in the atomic oxygen band and the auxiliary peak band other than the target band; denotes the final oxidation degree of the reference forging; denotes the oxidation degree of the target band; denotes the oxidation degree of another band in the atomic oxygen band and the auxiliary peak band other than the target band.

2. The method for spectral analysis for forging quality detection according to claim 1, characterized in that, The method for obtaining all wave bands on the spectral data curve comprises: Optionally, one forging spectral data in the reference forging is taken as reference data; The data difference between the reference data and the forging spectral data corresponding to the previous wavelength is calculated; All extreme points of the spectral data curve corresponding to the reference forging are obtained; The curve segment between each adjacent two extreme points is taken as an initial wave band; The peak value of the initial wave band in which the forging spectral data corresponding to the previous wavelength of the reference data is located is taken as a temporary wave peak; According to the position distribution of the reference data and the temporary wave peak and the data difference between the reference data and the forging spectral data corresponding to the previous wavelength, the possibility that the reference data and the previous forging spectral data are divided into the same initial wave band is obtained, and the formula is as follows: In the formula, represents the possible degree to which the first forging spectral data and the second forging spectral data are divided into the same initial waveband; represents the distance between the first forging spectral data of the initial waveband and the temporary wave peak; represents the distance between the temporary wave peak and the second forging spectral data; represents the second forging spectral data and the second forging spectral data; represents an exponential function with a natural constant as the base number; represents a normalization function; when When it exceeds the preset first threshold, the first... The spectral data of the forging part were divided into the first... The initial band where the spectral data of each forging is located, when When it is less than the preset first threshold, the first... The spectral data of one forging is used as the starting spectral data of the next band; all spectral data of forging are traversed, and the spectral data of all forging are divided into bands to obtain all bands on the spectral data curve.

3. The method for spectral analysis for forging quality detection according to claim 1, characterized in that, The method for obtaining the oxidation degree comprises: The oxidation degree is obtained according to the oxidation degree calculation formula, and the oxidation degree calculation formula is as follows: wherein represents the ordinal number of the target wavelength band; represents the oxidation degree of the target wavelength band; represents the peak value of the target wavelength band of the reference forging; represents the peak value of the target wavelength band of the forging in the normal state; represents the difference between the peak value of the target wavelength band of the forging in the normal state and the corresponding wavelength of the reference forging; represents a normalization function.

4. The method for spectral analysis for forging quality detection according to claim 1, characterized in that, The method for obtaining the oxidation weight comprises: The oxidation weight is obtained according to the oxidation weight calculation formula, and the oxidation weight calculation formula is as follows: In the formula, represents the serial number of the target waveband; represents the number of wavelengths in the target waveband in which the normal forging and the reference forging overlap; represents the forging spectral data corresponding to the first wavelength in which the normal forging overlaps in the target waveband; represents the forging spectral data corresponding to the first wavelength in which the reference forging overlaps in the target waveband; represents a normalization function.

5. The method for spectral analysis for forging quality detection according to claim 1, characterized in that, The method for obtaining the final spectral data comprises: The final spectral data is obtained according to the final spectral data calculation formula, and the final spectral data calculation formula is as follows: wherein represents the corresponding final spectral data of the reference forging in the th component; th forging spectral data of the th component of the reference forging; th forging spectral data of the th component of the reference forging; represents the change state of the forging spectral data corresponding to the th component; represents the final oxidation degree of the reference forging.

6. A spectroscopic analysis system for forging quality detection, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to realize the steps of the spectral analysis method for forging quality detection according to any one of claims 1 to 5.

7. A computer-readable storage medium storing a computer program, wherein the computer program comprises the following steps of: The computer program is executed by the processor to realize the steps of the spectral analysis method for forging quality detection according to any one of claims 1 to 5.

8. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor implements the steps of the spectrum analysis method for forging quality detection according to any one of claims 1 to 5 when executing the computer program.

Images